Three-dimensional data decoding method, three-dimensional data encoding device, and three-dimensional data decoding device

This is a continuation application of PCT International Application No. PCT/JP2022/022390 filed on Jun. 1, 2022, designating the United States of America, which is based on and claims priority of U.S. Provisional Patent Application No. 63/208,264 filed on Jun. 8, 2021 and U.S. Provisional Patent Application No. 63/209,111 filed on Jun. 10, 2021. The entire disclosures of the above-identified applications, including the specifications, drawings and claims are incorporated herein by reference in their entirety.

The present disclosure relates to a three-dimensional data encoding method, a three-dimensional data decoding method, a three-dimensional data encoding device, and a three-dimensional data decoding device.

Devices or services utilizing three-dimensional data are expected to find their widespread use in a wide range of fields, such as computer vision that enables autonomous operations of cars or robots, map information, monitoring, infrastructure inspection, and video distribution. Three-dimensional data is obtained through various means including a distance sensor such as a rangefinder, as well as a stereo camera and a combination of a plurality of monocular cameras.

Methods of representing three-dimensional data include a method known as a point cloud scheme that represents the shape of a three-dimensional structure by a point cloud in a three-dimensional space. In the point cloud scheme, the positions and colors of a point cloud are stored. While point cloud is expected to be a mainstream method of representing three-dimensional data, a massive amount of data of a point cloud necessitates compression of the amount of three-dimensional data by encoding for accumulation and transmission, as in the case of a two-dimensional moving picture (examples include Moving Picture Experts Group-4 Advanced Video Coding (MPEG-4 AVC) and High Efficiency Video Coding (HEVC) standardized by MPEG).

Meanwhile, point cloud compression is partially supported by, for example, an open-source library (Point Cloud Library) for point cloud-related processing.

Furthermore, a technique for searching for and displaying a facility located in the surroundings of the vehicle by using three-dimensional map data is known (see, for example, Patent Literature (PTL) 1).

PTL 1: International Publication WO 2014/020663

There has been a demand for improving coding efficiency in a three-dimensional data encoding process and a three-dimensional data decoding process.

The present disclosure has an object to provide a three-dimensional data encoding method, a three-dimensional data decoding method, a three-dimensional data encoding device, or a three-dimensional data decoding device that is capable of improving coding efficiency.

A three-dimensional data encoding method according to one aspect of the present disclosure comprising: determining, from among reference positions, a reference position corresponding to each of three-dimensional points; generating first information items each of which corresponds to a different one of the reference positions and indicates whether a three-dimensional point corresponding to a reference position corresponding to the first information item is present; selecting, from among contexts, a first context for a first information item to be processed that is included in the first information items; and arithmetic-encoding the first information item to be processed, using the first context.

A three-dimensional data decoding method according to one aspect of the present disclosure comprising: selecting, from among contexts, a first context for an encoded first information item to be processed; and arithmetic-decoding the encoded first information item to be processed, using the first context, to generate a first information item to be processed, wherein encoded first information items including the encoded first information item to be processed are included in a bitstream generated by encoding an information item about positions of three-dimensional points, the encoded first information items are generated by arithmetic-encoding first information items including the first information item to be processed, and the first information items each correspond to a different one of reference positions and indicate whether a three-dimensional point corresponding to a reference position corresponding to the first information item is present.

The present disclosure provides a three-dimensional data encoding method, a three-dimensional data decoding method, a three-dimensional data encoding device, or a three-dimensional data decoding device that is capable of improving coding efficiency.

A three-dimensional data encoding method according to one aspect of the present disclosure includes: determining, from among reference positions, a reference position corresponding to each of three-dimensional points; generating first information items each of which corresponds to a different one of the reference positions and indicates whether a three-dimensional point corresponding to a reference position corresponding to the first information item is present; selecting, from among contexts, a first context for a first information item to be processed that is included in the first information items; and arithmetic-encoding the first information item to be processed, using the first context.

Accordingly, the three-dimensional data encoding method is capable of arithmetic-encoding a first information item on a target point, using the first context selected from among the contexts. Accordingly, for example, a context appropriate to each first information item can be used. Therefore, coding efficiency can be improved.

For example, in the selecting of the first context, the first context may be selected based on an information item about a reference position corresponding to an encoded three-dimensional point among the reference positions.

For example, the information item about the reference position corresponding to the encoded three-dimensional point may be a first information item corresponding to the reference position corresponding to the encoded three-dimensional point.

Accordingly, the three-dimensional data encoding method is capable of appropriately selecting a context based on the first information item corresponding to the reference position corresponding to the encoded three-dimensional point. Here, a value of the first information item has a high correlation with values of other first information items in a predetermined range in at least one of a temporal direction and a spatial direction. That is, there is a high possibility that, in the vicinity of the first information item, a first information item having the same value appears in the at least one of the temporal direction or the spatial direction. Therefore, by selecting a context, using a first information item corresponding to a reference position corresponding to an encoded three-dimensional point, a context can be used with consideration given to correlation. Thus, it may be possible to reduce an encoding amount.

For example, the information item about the positions of the three-dimensional points may express each of the positions using a distance component, a first direction component, and a second direction component.

Accordingly, in the three-dimensional data encoding method, coding efficiency can be improved in encoding an information item about a position that is expressed with the distance component, the first direction component, and the second direction component.

For example, each of the reference positions may contain the first direction component and the second direction component, and in the selecting of the first context, the first context may be selected based on an information item about a first reference position that corresponds to an encoded three-dimensional point and contains a first direction component having a same value as a first direction component contained in a second reference position corresponding to the first information item to be processed.

Accordingly, in the three-dimensional data encoding method, the first context can be selected appropriately by referring to an information item about a reference position that includes a first direction component having the same value as a first direction component included in a reference position to be processed. Here, in what is called polar coordinates, when values of first direction components are assumed to be the same, values of first information items have high correlations even when values of second direction components are different. Therefore, by selecting a context, referring to the information item about the reference position that includes the first direction component having the same value as the first direction component included in the reference position to be processed, a context can be used with consideration given to correlation. Thus, it may be possible to reduce an encoding amount.

For example, first information items corresponding to reference positions may be arithmetic-encoded in an order based on the first direction component in a processing unit composed of the first information items, the reference positions containing second direction components having a same value.

Accordingly, by arithmetic-encoding first information items in a determined order, it may be possible that the same value is more likely to appear in processed first information items compared with the case where the first information items are arithmetic-encoded in no particular order. Accordingly, it may be possible to reduce a code amount in arithmetic encoding using a context.

For example, the first direction component may be a vertical component, the second direction component may be a horizontal component, and the first information items may be arithmetic-encoded in the order based on the vertical component.

Accordingly, there is continuity in presence or absence of points to be processed in the vertical direction. It may be possible to reduce an encoding amount in arithmetic encoding.

For example, each of the reference positions may contain the first direction component and the second direction component, and the three-dimensional data encoding method may further comprise: calculating a first difference between a value of the first direction component of a three-dimensional point to be processed and a value of the first direction component contained in a first reference position corresponding to the three-dimensional point to be processed.

Accordingly, in the three-dimensional data encoding method, coding efficiency can be improved by calculating the first difference between the value of the first direction component of the three-dimensional point to be processed and the value of the first direction component of the first reference position.

For example, the three-dimensional data encoding method may further comprise: calculating a second difference between a value of the first direction component of an encoded three-dimensional point and a value of the first direction component contained in a second reference position corresponding to the encoded three-dimensional point; and calculating a third difference that is a difference between the first difference and the second difference.

Accordingly, in the three-dimensional data encoding method, coding efficiency can be further improved.

For example, the reference positions may correspond to sampling positions in a sensor for generating the three-dimensional points.

For example, when intra prediction is applied to a three-dimensional point to be processed, the first context may be selected by a first method, and when inter prediction is applied to the three-dimensional point to be processed, the first context may be selected by a second method that is different from the first method.

Accordingly, in the three-dimensional data encoding method, a context appropriate to each of the intra prediction and the inter prediction can be selected. Thus, coding efficiency can be improved.

For example, in the second method, the first context may be selected by reference to a processed frame that is different from a frame to be processed that includes the three-dimensional point to be processed.

Accordingly, in the three-dimensional data encoding method, a context appropriate to the inter prediction can be selected. Thus, coding efficiency can be improved.

A three-dimensional data decoding method according to one aspect of the present disclosure includes: selecting, from among contexts, a first context for an encoded first information item to be processed; and arithmetic-decoding the encoded first information item to be processed, using the first context, to generate a first information item to be processed, wherein encoded first information items including the encoded first information item to be processed are included in a bitstream generated by encoding an information item about positions of three-dimensional points, the encoded first information items are generated by arithmetic-encoding first information items including the first information item to be processed, and the first information items each correspond to a different one of reference positions and indicate whether a three-dimensional point corresponding to a reference position corresponding to the first information item is present.

Accordingly, the three-dimensional data decoding method is capable of arithmetic-decoding a first information item on a target point, using the first context selected from among the contexts. Accordingly, for example, a context appropriate to each first information item can be used. Therefore, the three-dimensional data decoding method is capable of appropriately arithmetic-decoding a bitstream of which coding efficiency is improved.

For example, in the selecting of the first context, the first context may be selected based on an information item about a reference position corresponding to a decoded three-dimensional point among the reference positions.

For example, the information item about the reference position corresponding to the decoded three-dimensional point may be a first information item corresponding to the reference position corresponding to the decoded three-dimensional point.

Accordingly, the three-dimensional data decoding method is capable of appropriately selecting a context based on the first information item corresponding to the reference position corresponding to the decoded three-dimensional point. Here, a value of the first information item has a high correlation with values of other first information items in a predetermined range in at least one of a temporal direction and a spatial direction. That is, there is a high possibility that, in the vicinity of the first information item, a first information item having the same value appears in the at least one of the temporal direction or the spatial direction. Therefore, by selecting a context, using a first information item corresponding to a reference position corresponding to a decoded three-dimensional point, a context can be used with consideration given to correlation. Thus, it may be possible to reduce an encoding amount.

For example, the information item about the positions of the three-dimensional points may express each of the positions using a distance component, a first direction component, and a second direction component.

Accordingly, in the three-dimensional data decoding method, an information item about a position that is expressed with the distance component, the first direction component, and the second direction component can be decoded from a bitstream of which coding efficiency is improved.

For example, each of the reference positions may contain the first direction component and the second direction component, and in the selecting of the first context, the first context may be selected based on an information item about a first reference position that corresponds to a decoded three-dimensional point and contains a first direction component having a same value as a first direction component contained in a second reference position corresponding to the first information item to be processed.

Accordingly, in the three-dimensional data decoding method, the first context can be selected appropriately by referring to an information item about a reference position that includes a first direction component having the same value as a first direction component included in a reference position to be processed. Here, in what is called polar coordinates, when values of first direction components are assumed to be the same, values of first information items have high correlations even when values of second direction components are different. Therefore, by selecting a context, referring to the information item about the reference position that includes the first direction component having the same value as the first direction component included in the reference position to be processed, a context can be used with consideration given to correlation. Thus, it may be possible to reduce an encoding amount.

For example, first information items corresponding to reference positions may be arithmetic-decoded in an order based on the first direction component in a processing unit composed of the first information items, the reference positions containing second direction components having a same value.

Accordingly, by arithmetic-decoding first information items in a determined order, it may be possible that the same value is more likely to appear in processed first information items compared with the case where the first information items are arithmetic-decoded in no particular order. Accordingly, it may be possible to reduce a code amount.

For example, the first direction component may be a vertical component, the second direction component may be a horizontal component, and the first information items may be arithmetic-decoded in the order based on the vertical component.

Accordingly, there is continuity in presence or absence of points to be processed in the vertical direction. It may be possible to reduce an encoding amount.

For example, each of the reference positions may contain the first direction component and the second direction component, and the first information item to be processed may indicate a first difference between a value of the first direction component of a three-dimensional point to be processed and a value of the first direction component contained in a first reference position corresponding to the three-dimensional point to be processed.

Accordingly, in the three-dimensional data decoding method, a bitstream of which coding efficiency is improved can be appropriately decoded.

For example, each of the reference positions may contain the first direction component and the second direction component, the first information item to be processed may indicate a third difference that is a difference between a first difference and a second difference, the first difference may be a difference between a value of the first direction component of the three-dimensional point to be processed and a value of the first direction component contained in a first reference position corresponding to the three-dimensional point to be processed, and the second difference may be a difference between a value of the first direction component of a decoded three-dimensional point and a value of the first direction component contained in a second reference position corresponding to the decoded three-dimensional point.

Accordingly, in the three-dimensional data decoding method, a bitstream of which coding efficiency is improved can be appropriately decoded.

For example, the reference positions may correspond to sampling positions in a sensor for generating the three-dimensional points.

For example, when intra prediction is applied to a three-dimensional point to be processed, the first context may be selected by a first method, and when inter prediction is applied to the three-dimensional point to be processed, the first context may be selected by a second method that is different from the first method.

Accordingly, in the three-dimensional data decoding method, a context appropriate to each of the intra prediction and the inter prediction can be selected. Thus, coding efficiency can be improved.

For example, in the second method, the first context may be selected by reference to a processed frame that is different from a frame to be processed that includes the three-dimensional point to be processed.

Accordingly, in the three-dimensional data decoding method, a context appropriate to the inter prediction can be selected. Thus, coding efficiency can be improved.

A three-dimensional data encoding device according to one aspect of the present disclosure includes a processor and memory. Using the memory, the processor determines, from among reference positions, a reference position corresponding to each of three-dimensional points; generates first information items each of which corresponds to a different one of the reference positions and indicates whether a three-dimensional point corresponding to a reference position corresponding to the first information item is present; selects, from among contexts, a first context for a first information item to be processed that is included in the first information items; and arithmetic-encodes the first information item to be processed, using the first context.

Accordingly, the three-dimensional data encoding device is capable of arithmetic-encoding a first information item on a target point, using the first context selected from among the contexts. Accordingly, for example, a context appropriate to each first information item can be used. Therefore, coding efficiency can be improved.

A three-dimensional data decoding device according to one aspect of the present disclosure includes a processor and memory. Using the memory, the processor selects, from among contexts, a first context for an encoded first information item to be processed; and arithmetic-decodes the encoded first information item to be processed, using the first context, to generate a first information item to be processed. Encoded first information items including the encoded first information item to be processed are included in a bitstream generated by encoding an information item about positions of three-dimensional points. The encoded first information items are generated by arithmetic-encoding first information items including the first information item to be processed. The first information items each correspond to a different one of reference positions and indicate whether a three-dimensional point corresponding to a reference position corresponding to the first information item is present.

Accordingly, the three-dimensional data decoding device is capable of arithmetic-decoding a first information item on a target point, using the first context selected from among the contexts. Accordingly, for example, a context appropriate to each first information item can be used. Therefore, the three-dimensional data decoding device is capable of appropriately arithmetic-decoding a bitstream of which coding efficiency is improved.

It is to be noted that these general or specific aspects may be implemented as a system, a method, an integrated circuit, a computer program, or a computer-readable recording medium such as a CD-ROM, or may be implemented as any combination of a system, a method, an integrated circuit, a computer program, and a recording medium.

Hereinafter, embodiments will be specifically described with reference to the drawings. It is to be noted that each of the following embodiments indicate a specific example of the present disclosure. The numerical values, shapes, materials, constituent elements, the arrangement and connection of the constituent elements, steps, the processing order of the steps, etc., indicated in the following embodiments are mere examples, and thus are not intended to limit the present disclosure. Among the constituent elements described in the following embodiments, constituent elements not recited in any one of the independent claims will be described as optional constituent elements.

100 100 100 1 FIG. First, a configuration of three-dimensional data encoding deviceaccording to the present embodiment will be described.is a block diagram of three-dimensional data encoding deviceaccording to the present embodiment. Three-dimensional data encoding devicegenerates a bitstream (encoded stream) by encoding point cloud data being three-dimensional data.

The point cloud data includes position information items on three-dimensional points. The position information items indicate three-dimensional positions of the three-dimensional points. It should be noted that the position information items may be also referred to as geometry information items.

For example, the position information items are each expressed in a polar coordinate system and each include one distance component and two direction components (angle components). Specifically, the position information items each include distance d, elevation angle θ, and horizontal angle φ. The point cloud data is, for example, data obtained by a laser sensor such as LiDAR.

1 FIG. 100 The point cloud data may include attribute information items (color, reflectivity, etc.) of the three-dimensional points in addition to the position information items.illustrates a processor unit relating to encoding of the position information items on the point cloud data. However, three-dimensional data encoding devicemay include another processor unit such as a processor unit that performs encoding and the like of the attribute information items.

100 101 102 103 104 105 106 108 109 110 111 112 113 Three-dimensional data encoding deviceincludes transformer, subtractor, quantizer, entropy encoder, inverse quantizer, adder, buffer, intra predictor, buffer, motion detector and compensator, inter predictor, and switcher.

101 101 101 101 101 Transformergenerates transform information items by transforming position information items included in point cloud data to be encoded that is input. Specifically, transformergenerates information for associating reference positions and three-dimensional points with each other. Transformeralso transforms position information items on the three-dimensional points, using the reference positions. For example, the transform information items include differences between the reference positions and the position information items on the three-dimensional points. This will be described later in detail. Transformermay include a buffer for retaining the transformed position information items. In other words, transformeris a calculator that calculates a value to be encoded.

102 103 104 104 101 Subtractorgenerates residual signals (also referred to as prediction residuals) by subtracting predicted values from the transformed position information items. Quantizerquantizes the residual signals. Entropy encodergenerates a bitstream by entropy encoding the quantized residual signals. Entropy encoderalso entropy encodes control information that is information and the like generated by transformerand adds the encoded information to the bitstream.

105 103 106 105 108 110 Inverse quantizergenerates residual signals by inverse quantizing the quantized residual signals obtained by quantizer. Adderreproduces transform information items by adding predicted values to the residual signals generated by inverse quantizer. Bufferretains the reproduced transform information items as a reference point cloud for intra prediction. Bufferretains the reproduced transform information items as a reference point cloud for inter prediction.

It should be noted that the reproduced transform information items each include a quantization error and thus may fail to perfectly match original transform information items. Three-dimensional points that are reproduced by an encoding process and a decoding process in this manner will be referred to as encoded three-dimensional points, decoded three-dimensional points, or processed three-dimensional points.

109 Intra predictorcalculates a predicted value, using transform information items on one or a plurality of reference points other than a three-dimensional point to be processed (hereinafter referred to as a target point), the reference points being processed three-dimensional points belonging to the same frame as a frame to which the target point belongs.

111 Motion detector and compensatordetects a displacement between a target frame being a frame including the target point and a reference frame being a frame different from the target frame (motion detection) and corrects transform information items on a point cloud included in the reference frame based on the detected displacement (motion compensation). Information indicating the detected displacement (motion information) is stored in, for example, the bitstream.

112 Inter predictorcalculates a predicted value, using transform information items on one or a plurality of reference points included in the point cloud that are subjected to the motion compensation. It should be noted that the motion detection and the motion compensation need not be performed.

113 109 112 102 106 113 113 Switcherselects one of the predicted value calculated by intra predictorand the predicted value calculated by inter predictorand outputs the selected predicted value to subtractorand adder. That is, switcherswitches whether to apply the intra prediction or the inter prediction. For example, switchercalculates a cost value of the case where the intra prediction is applied and a cost value of the case where the inter prediction is applied and selects a prediction method that results in a smaller cost value. The cost value is, for example, a value based on a code amount after encoding. A smaller code amount results in a smaller cost value. In the case where the intra prediction and the inter prediction each have a plurality of techniques (a plurality of prediction modes), a prediction mode to be used is determined based on the cost value in the same manner. It should be noted that a method of determining the prediction method (the intra prediction or inter prediction) and the prediction mode is not limited to this. The prediction method and the prediction mode may be determined based on settings designated from the outside or properties of the point cloud data. Alternatively, selectable candidates of the prediction method and selectable candidates of the prediction mode may be narrowed down based on settings designated from the outside or properties of the point cloud data.

100 100 101 100 106 101 Three-dimensional data encoding devicemay obtain position information items expressed in a Cartesian coordinate system, transform the position information items in the Cartesian coordinate system to position information items in a polar coordinate system, and perform the encoding process on the obtained position information items in the polar coordinate system. For example, three-dimensional data encoding devicemay include a coordinate transformer performing the coordinate transformation process in a stage prior to transformer. In this case, three-dimensional data encoding devicemay generate position information items in the polar coordinate system by performing, on the transform information items reproduced by adder, inverse transform of the transform process performed by transformer, transform the generated position information items in the polar coordinate system to position information items in the Cartesian coordinate system, calculate differences between the obtained position information items in the Cartesian coordinate system and original position information items in the Cartesian coordinate system being input, and store information indicating the calculated differences in a bitstream.

200 100 200 200 200 100 2 FIG. 2 FIG. 1 FIG. Next, a configuration of three-dimensional data decoding devicethat decodes a bitstream generated by three-dimensional data encoding devicedescribed above will be described.is a block diagram of three-dimensional data decoding deviceaccording to the present embodiment.illustrates a processor unit relating to decoding of position information items on a point cloud. However, three-dimensional data decoding devicemay include another processor unit such as a processor unit that performs decoding and the like of attribute information items on the point cloud. For example, three-dimensional data decoding devicegenerates decoded point cloud data by decoding a bitstream generated by three-dimensional data encoding deviceillustrated in.

200 201 202 203 204 205 206 207 208 209 210 Three-dimensional data decoding deviceincludes entropy decoder, inverse quantizer, adder, inverse transformer, buffer, intra predictor, buffer, motion compensator, inter predictor, and switcher.

200 100 Three-dimensional data decoding deviceobtains a bitstream generated by three-dimensional data encoding device.

201 Entropy decoderentropy decodes the bitstream, thus generating quantized residual signals, control information, and the like.

202 201 203 202 Inverse quantizergenerates residual signals by inverse quantizing the quantized residual signals obtained by entropy decoder. Adderreproduces transform information items by adding predicted values to the residual signals generated by inverse quantizer.

204 101 204 204 204 204 Inverse transformerreproduces position information items by performing, on the transform information items, inverse transform of the transform process performed by transformer. Specifically, inverse transformerobtains, from the bitstream, information for associating reference positions and three-dimensional points with each other and associates the reference positions and the three-dimensional points with each other based on the obtained information. Inverse transformeralso transforms position information items on the three-dimensional points, using the reference positions. For example, inverse transformercalculates the position information items by adding the transform information items and the reference positions together. In other words, inverse transformeris a calculator that calculates a position information item from a decoded value. The position information items are output as the decoded point cloud data.

205 203 207 203 206 Bufferretains the transform information items reproduced by adderas a reference point cloud for the intra prediction. Bufferretains the transform information items reproduced by adderas a reference point cloud for the inter prediction. Intra predictorcalculates a predicted value, using transform information items on one or a plurality of reference points other than a target point, the reference points being processed three-dimensional points belonging to the same frame as a frame to which the target point belongs.

208 209 Motion compensatorobtains, from the bitstream, motion information indicating a displacement between a target frame and a reference frame and corrects, based on the displacement indicated by the motion information, transform information items on a point cloud included in the reference frame (motion compensation). Inter predictorcalculates a predicted value, using transform information items on one or a plurality of reference points included in the point cloud that are subjected to the motion compensation. It should be noted that the motion compensation need not be performed.

210 206 209 203 210 Switcherselects one of the predicted value calculated by intra predictorand the predicted value calculated by inter predictorand outputs the selected predicted value to adder. For example, switcherobtains information indicating the prediction method (the intra prediction or the inter prediction) from the bitstream and determines a prediction method to be used based on the obtained information. In the case where the intra prediction and the inter prediction each have techniques (prediction modes), information indicating a prediction mode is obtained from a bitstream, and a prediction mode to be used is determined based on the obtained information, in the same manner.

200 200 204 200 200 204 Three-dimensional data decoding devicemay transform decoded position information items expressed in the polar coordinate system to position information items expressed in the Cartesian coordinate system and output the position information items expressed in the Cartesian coordinate system. For example, three-dimensional data decoding devicemay include a coordinate transformer performing the coordinate transformation process in a stage subsequent to inverse transformer. In this case, three-dimensional data decoding deviceobtains, from the bitstream, information indicating differences between original position information items in the Cartesian coordinate system before being encoded and decoded and the decoded position information items in the Cartesian coordinate system. Three-dimensional data decoding devicemay transform the position information items in the polar coordinate system reproduced by inverse transformerto position information items in the Cartesian coordinate system, add the differences indicated by the information to the obtained position information items in the Cartesian coordinate system, and output the obtained position information items in the Cartesian coordinate system.

100 101 3 FIG. 3 FIG. Next, operation of three-dimensional data encoding devicewill be described.is a diagram illustrating operation of transformer.is a diagram illustrating an encoding order (processing order) of three-dimensional points (reference positions) in the encoding process.

3 FIG. 3 FIG. 3 FIG. 101 In, a horizontal direction represents horizontal angle φ in polar coordinates, and a vertical direction represents elevation angle θ in the polar coordinates. Transformersets reference positions rm (m=0, 1, 2, . . . ) (also referred to as reference points). Here, reference positions rm are each expressed with horizontal angle φ and elevation angle θ. In other words, reference positions rm are each expressed with two components (θ, φ) out of three components (d, θ, φ) that express a position information item on a three-dimensional point. In addition, in the example illustrated in, reference positions rm indicated by squares in the figure are set based on sampling interval Δφ that is a horizontal sampling interval of LiDAR and scan-line interval Δθk of LiDAR (k=1, 2, 3). In other words, the reference positions are set based on combinations of predetermined horizontal angles and elevation angles and disposed on a plane expressed by horizontal angle φ and elevation angle θ in a matrix pattern. In the example illustrated in, intervals Δφ between horizontal angles φj (j=0, 1, 2, . . . ) of the reference positions are constant. Intervals between elevation angles θk (k=0, 1, 2, 3) of the reference positions can be set individually.

101 Transformerperforms an encoding process (a transform process) on points pn (n=0, 1, 2, . . . ) indicated by rhombi located in the vicinities of the reference positions in an order indicated by dashed arrows in the figure. Hatched squares indicate first reference positions where points referring to the reference positions are present, and squares not hatched indicate second reference positions where points referring to the reference positions are not present.

The points referring to the reference position are points based on the reference positions. The points are associated with the reference positions (encoded (transformed) using the reference positions) as will be described later. In addition, the points referring to the reference positions are each a point of which values of horizontal angle φ and elevation angle θ are included within their respective ranges including the corresponding reference position. For example, the points referring to the reference positions are points pn that have horizontal angles being greater than or equal to φj and less than φj+Δφ and are on the same scan line (have the same elevation angle). The range in horizontal angle is not limited to this. The range in horizontal angle may be, for example, greater than or equal to φj−Δφ/2 and less than φj+Δφ/2.

3 FIG. 3 FIG. 3 FIG. 3 FIG. 3 FIG. The processing order (encoding order) illustrated inis based on processing units (corresponding to columns in) each consisting of reference positions having horizontal angles of the same value (e.g., r0 to r3), and in each processing unit, the reference positions are processed (encoded) in an order based on the elevation angle (ascending order in). The processing units (corresponding to the columns in) are processed in an order based on the horizontal angle (ascending order in). In other words, the reference positions are processed in ascending order of elevation angle for each set of reference positions having horizontal angles of the same value. The reference positions may be processed in ascending order of horizontal angle for each set of reference positions having elevation angles of the same value.

101 101 o o o o In encoding (transforming) of a target point, transformergenerates information for identifying a position (φj, θk) of reference position rm that is referred to by target point pn. Transformergenerates an offset (φn, θn) from the reference position to the target point and information for identifying distance information dn on the target point. Here, φn is a difference between horizontal angle φj of the reference position and a horizontal angle of the target point, and θn is a difference between elevation angle θk of the reference position and an elevation angle of the target point.

o o The information for identifying the position of the reference position that is referred to by the target point, offset (φn, θn) from the reference position to the target point, and the information for identifying distance information dn on the target point each may be information for identifying a difference value from a predicted value generated based on processed information or may be information for identifying the value itself.

100 200 Three-dimensional data encoding devicemay also store sampling interval Δφ that is a horizontal sampling interval of LiDAR and scan-line interval Δθk of LiDAR in a bitstream. For example, the three-dimensional data encoding device may store Δφ and Δθk in a header of an SPS or a GPS. Accordingly, three-dimensional data decoding devicecan set the reference positions, using Δφ and Δθk.

Here, the SPS (sequence parameter set) is a sequence-based parameter set (control information) where a sequence includes frames. The SPS is also a parameter set common to geometry information and attribute information. The GPS (geometry parameter set) is a frame-based parameter set. The GPS is a parameter set for geometry information.

101 3 FIG. The transformermay also transform sampling interval Δφ being a horizontal sampling interval of LiDAR and scan-line interval Δθk of LiDAR to values that are rounded to integer values having a predetermined bit width, and store the transformed values in a bitstream. The example illustrated inis an example in which the number of scan lines (the number of elevation angles) is 4 but can be carried out in the case where another number of scan lines such as 16, 64, or 128 is used.

4 FIG. 4 FIG. 5 FIG. 100 Next, syntax of the geometry information will be described.is a diagram illustrating an example of syntax of a geometry information item on each point. In syntax examples shown inand, parameters (signals) stored in a bitstream are written in bold type. Three-dimensional data encoding devicerepeatedly applies this syntax for each reference position rm to generate column_pos, which indicates an index of horizontal angle φj of reference position rm serving as a reference for point pn to be processed next, and row_pos, which indicates an index of elevation angle θk of reference position rm, and further generates parameter relating to point pn.

100 100 100 200 100 200 In this example, three-dimensional data encoding deviceinitializes variables before processing a first point. Specifically, three-dimensional data encoding devicesets first_point_in_column, which indicates a first piece of syntax corresponding to horizontal angles φj, to 1, sets column_pos to 0, and sets row_pos to 0. Alternatively, three-dimensional data encoding devicemay notify three-dimensional data decoding deviceof a value of column_pos and a value of row_pos of the first point, in advance of syntax corresponding to the first point. In this case, three-dimensional data encoding deviceand three-dimensional data decoding devicemay apply this syntax, using these values after setting first_point_in_column to 0.

100 3 FIG. 3 FIG. 3 FIG. Next, three-dimensional data encoding devicegenerates next_column_flag at reference position rm corresponding to a position having an elevation angle being θ0 (i.e., in the case where first_point_in_column is 1). next_column_flag indicates whether there is one or more points based on horizontal angles φj corresponding to the position of reference position rm. In other words, next_column_flag indicates whether there is a point that refers to any one of reference positions having the same horizontal angle as horizontal angles φj of reference position rm. For example, in the case where there is one or more points based on horizontal angle φj corresponding to the position of reference position rm (e.g., horizontal angles φ0, φ1, φ2, and φ4 illustrated in), next_column_flag is set to 0, and in the case where there is no point based on horizontal angle φj corresponding to the position of reference position rm (e.g., horizontal angle φ3 illustrated in), next_column_flag is set to 1. next_column_flag is provided for each horizontal angle φj (for each column in).

100 By repeatedly generating next_column_flag until next_column_flag becomes 0, three-dimensional data encoding devicecan generate information that enables identification of horizontal angle φj corresponding to point pn to be processed next (φ0+column_pos×Δφ). Accordingly, it may be possible to reduce a code amount required to notify next_row_flag described below.

5 FIG. Whether to notify next_column_flag can be determined by whether row_pos is 0, as will be shown indescribed later. However, determination with first_point_in_column enables avoidance with the notification of next_column_flag, which is also unnecessary in the case where there are points at a position of row_pos being 0, and thus can reduce the code amount.

100 3 FIG. 3 FIG. Three-dimensional data encoding devicegenerates next_row_flag at each candidate position of reference position rm serving as a reference for point pn to be processed next. next_row_flag indicates whether there is point pn to be processed at a position of elevation angle θk. In other words, next_row_flag indicates whether there is a point that refers to reference position rm. For example, when there is point pn to be processed at a position of elevation angle θk, next_row_flag is set to 0 (e.g., r0 and r1 in), and when there is no point pn to be processed at a position of elevation angle θk (e.g., r2 and r3 in), next_row_flag is set to 1. next_row_flag is provided for each reference position.

100 100 4 FIG. When next_row_flag is 1, three-dimensional data encoding devicerepeatedly applies the syntax illustrated into generate next_row_flag corresponding to each candidate position consecutively. By repeating this process until next_row_flag becomes 0, three-dimensional data encoding devicecan generate information that enables identification of elevation angle θk corresponding to point pn to be processed next. For example, elevation angle θk corresponding to point pn to be processed next is given by (Expression 1) shown below.

4 FIG. 100 When row_pos reaches the number of scan lines (num_rows illustrated in), the process proceeds to next horizontal angle φj. At this time, three-dimensional data encoding devicesets row_pos to 0, increases column_pos by 1, and sets first_point_in_column to 1.

100 In the above-described manner, three-dimensional data encoding devicecan generate the information items (next_column_flag and next_row_flag) that enable the identification of horizontal angle φj and elevation angle θk of reference position rm serving as the reference for point pn to be processed.

100 o Subsequently, three-dimensional data encoding devicegenerates information relating to a distance of target point pn, information relating to an offset in horizontal angle from reference position rm to target point pn, and pred_mode, which is information relating to a prediction method for these parameters. Here, the information relating to the distance is, for example, residual residual_radius, which indicates a difference between the distance of the target point and a predicted value generated by a predetermined method. The information relating to the offset in horizontal angle is, for example, residual residual_phi, which indicates a difference between offset φn in horizontal angle and a predicted value generated by a predetermined method.

100 100 The predicted values are calculated based on, for example, information on a processed three-dimensional point. For example, the predicted values are at least some of parameters of one or more processed three-dimensional points located in the vicinity of the target point. In this example, three-dimensional data encoding deviceomits generation of information relating to an offset in elevation angle assuming that an offset in elevation angle is always 0. However, three-dimensional data encoding devicemay generate information relating to an offset in elevation angle from reference position rm to point pn to be processed and store the information in a bitstream. For example, the information relating to an offset in elevation angle is residual residual_theta, which indicates a difference between offset bon of an elevation angle and a predicted value generated by a predetermined method.

100 100 106 1 FIG. Three-dimensional data encoding devicemay transform position information items in a Cartesian coordinate system being input to position information items expressed in a polar coordinate system, and perform the encoding process on the obtained position information items expressed in the polar coordinate system. In this case, three-dimensional data encoding devicemay transform again position information items in the polar coordinate system encoded and decoded (e.g., position information items generated by inverse transforming signals output from adderillustrated in) to position information items in the Cartesian coordinate system, calculate differences between the obtained position information items in the Cartesian coordinate system and original position information items in the Cartesian coordinate system being input, and store information indicating the differences in a bitstream. The information indicating the differences includes, for example, correction values residual_x, residual_y, and residual_z on X, Y, and Z axes, respectively. In other words, in the case where the transform between the coordinate systems is not performed, residual_x, residual_y, and residual_z need not be included in the bitstream.

200 104 next_column_flag, next_row_flag, pred_mode, residual_radius, residual_phi, residual_theta, residual_x, residual_y, and residual_z generated in the above-described manner are stored in a bitstream and sent to three-dimensional data decoding device. It should be noted that all or some of these signals may be entropy encoded (arithmetic-encoded) by entropy encoderbefore being stored in the bitstream.

100 As described above, three-dimensional data encoding devicecan determine values of syntax elements for each candidate position for reference position rm by using the information items for identifying horizontal angle φj and elevation angle θk of reference position rm serving as the reference for point pn to be processed next and the flags (next_column_flag and next_row_flag) associated with each candidate position. In addition, it may be possible to make the encoding process, the decoding process, a data send process, or the like less delayed.

It should be noted that application of values to the syntax including next_column_flag, next_row_flag, and the like as well as the variables such as first_point_in_column in the description is merely an example. For example, the application may be changed such that 0 and 1 are inversely applied. This case can be carried out by conforming related conditional determinations and the like.

5 FIG. 5 FIG. 4 FIG. 100 Next, another example of the syntax will be described.is a diagram illustrating an example of the syntax of a geometry information item on each point. Three-dimensional data encoding devicerepeatedly applies this syntax for each reference position rm to generate column_pos, which indicates an index of horizontal angle φj of reference position rm serving as a reference for point pn to be processed next, and row_pos, which indicates an index of elevation angle θk of reference position rm, and further generates parameter relating to point pn. The example shown indiffers from the example shown inin the method of generating next_row_flag and next_column_flag used for identifying the values of column_pos and row_pos.

100 100 200 100 100 200 In this example, three-dimensional data encoding devicefirst initializes variables before applying the syntax to a first point. Specifically, three-dimensional data encoding devicenotifies three-dimensional data decoding deviceof a value of column_pos and a value of row_pos of the first point, in advance of syntax corresponding to the first point. In other words, for example, three-dimensional data encoding devicestores the value of column_pos and the value of row_pos of the first point in a bitstream. Three-dimensional data encoding deviceand three-dimensional data decoding deviceapply the syntax with these values.

100 200 Next, three-dimensional data encoding devicegenerates next_row_flag for reference position rm at a position indicated by next_row_flag and next_column_flag and notifies three-dimensional data decoding devicewhether there is point pn based on reference position rm at the position.

100 100 100 100 100 100 5 FIG. 5 FIG. When next_row_flag is 1, three-dimensional data encoding devicefirst increases row_pos by 1. Next, three-dimensional data encoding devicedetermines whether row_pos has reached the number of scan lines (num_rows shown in). When row_pos reaches the number of scan lines, three-dimensional data encoding devicesets row_pos to 0 and increases column_pos by 1, determining that a candidate position is to be shifted to next horizontal angle φj. Next, three-dimensional data encoding devicedetermines whether row_pos is 0. When row_pos is 0, three-dimensional data encoding devicegenerates one or more next_column_flag and repeatedly increases column_pos by 1 until next_column_flag becomes 0. Thereafter, three-dimensional data encoding devicerepeatedly applies the syntax shown inuntil next_row_flag becomes 0.

100 5 FIG. 4 FIG. When next_row_flag is 0, three-dimensional data encoding devicedetermines the values indicated by next_row_flag and next_column_flag at the time to be an index of horizontal angle φj and an index of elevation angle θk of reference position rm serving as a reference for point pn to be processed next and stores parameters relating to point pn to be processed next (e.g., pred_mode, residual_radius, residual_phi, residual_x, residual_y, and residual_z shown in) in a bitstream as in the example shown in. Horizontal angle φj can be calculated by φ0+column_pos×Δφ, using values of the indices and sampling interval Δφ that is a horizontal sampling interval of LiDAR. Elevation angle θk can be calculated by (Expression 1) shown above, using the values of the indices and scan-line interval Δθk of LiDAR.

In the case where the transform between the coordinate systems is not performed, residual_x, residual_y, and residual_z need not be included in the bitstream. residual_theta may be included in the bitstream.

In the above-described manner, the notification of next_column_flag can be limited only to the case where row_pos=0 and next_row_flag=1, and thus it may be possible to reduce a code amount.

100 As described above, three-dimensional data encoding devicecan determine values of syntax elements for each candidate position for reference position rm by using the information items for identifying horizontal angle φj and elevation angle θk of reference position rm serving as the reference for point pn to be processed next and the flags (next_column_flag and next_row_flag) associated with each candidate position. In addition, it may be possible to make the encoding process, the decoding process, a data send process, or the like less delayed.

It should be noted that application of values to the syntax including next_column_flag, next_row_flag, and the like in the description is merely an example. For example, the application may be changed such that 0 and 1 are inversely applied. This case can be carried out by conforming related conditional determinations and the like.

6 FIG. 6 FIG. Next, an arithmetic encoding process of next_row_flag will be described.is a diagram for describing an example of a method for selecting a context (a probability table) in arithmetic encoding of next_row_flag. Reference position rm indicated inis a reference position corresponding to next_row_flag to be encoded.

104 104 104 6 FIG. Entropy encodercan use information items about reference positions included in a processed range indicated by shading surrounded by broken lines infor the encoding process on next_row_flag corresponding to reference position rm. For example, entropy encoderretains, in a memory, a predetermined number of one or more first reference positions at which points that are processed recently at each scan line and refer to the reference positions (e.g., hatched squares such as reference positions A0, B0, and C0 on a scan line of elevation angle θ0). Based on one or more information items about the one or more first reference positions retained in the memory, entropy encoderswitches among contexts used for arithmetic encoding of next_row_flag.

104 104 104 104 104 For example, entropy encoderuses an information item about at least one of reference positions A1, B1, and C1 that are located on the same scan line as reference position rm. Specifically, entropy encodermay use a difference in column_pos between at least one of reference positions A1, B1, and C1 and reference position rm. For example, entropy encodermay use a difference in column_pos between reference position A1 closest to reference position rm and reference position rm. Alternatively, entropy encodermay use a combination of the difference in column_pos between reference position A1 being closest to reference position rm and reference position rm and a difference in column_pos between reference position B1 being next closest to reference position rm and reference position rm. In this manner, entropy encodermay determine a context in accordance with whether one or more reference positions located on the same scan line as reference position rm are first reference positions (whether there are one or more points referring to the one or more reference positions). Here, in point cloud data obtained by LiDAR, for example, points located on the same scan line may have a high correlation. Therefore, by referring to information on the points located on the same scan line to select a context, the selection of a context can be performed appropriately.

104 104 104 Alternatively, entropy encodermay use an information item about a first reference position that is processed most recently (e.g., reference position A0). Specifically, entropy encodermay switch among contexts in accordance with the number of times next_row_flag is 1 consecutively from reference position A0 to reference position rm. Alternatively, entropy encodermay switch among contexts in accordance with row_pos of reference position rm itself rather than the information items about reference positions retained in the memory.

Switching among contexts is selecting a context to be used from among a plurality of contexts. Each of the contexts may be a context having a fixed probability or may be a context having an updatable probability. For the context having an updatable probability, the probability is updated in accordance with a value (0, 1) of an arithmetic-encoded signal, and in a subsequent arithmetic encoding process using the same context, the updated probability is used.

104 As described above, entropy encoderselects a context (a probability table) to be used in arithmetic encoding of next_row_flag using information items about processed reference positions. Accordingly, the contexts can be switched to one another finely, and it may be possible to improve coding efficiency.

6 FIG. 6 FIG. Next, a method for intra prediction of information on a distance and a horizontal angle will be described with reference to. An example of a method for predicting the information relating to the distance of target point pn and the information relating to the horizontal angle of target point pn will be described below. Here, reference position rm indicated inis a reference position that is referred to by target point pn. The intra prediction is a process of encoding a target point included in a target frame with reference to a reference point that is an encoded point different from the target point and included in the target frame. Specifically, one or more reference points are used to generate a predicted value, and a difference (residual) between information on the target point and the predicted value is calculated.

109 109 109 109 109 109 In an intra prediction process of target point pn, intra predictorcan use information items about reference positions and points within the processed range indicated by shading surrounded by the broken lines in the figure. For example, intra predictorretains, in a memory, a predetermined number of information items on points that are processed most recently on each scan line (e.g., hatched rhombi such as points a1, b1, and c1 on a scan line of elevation angle θ1). Intra predictormay predict information on target point pn based on the information items about the points retained in the memory. More specifically, intra predictorretains, in the memory, information relating to distances and horizontal angles of points a1, b1, and c1 located on the same scan line as target point pn. Intra predictordetermines a prediction mode to be used and calculates a predicted value, using the determined prediction mode. Intra predictorstores pred_mode indicating the determined prediction mode in a bitstream.

109 109 109 109 109 o Specifically, intra predictoruses a distance of a reference point as a predicted value of the distance of target point pn. Intra predictorcalculates a difference (residual component) between the predicted value and the distance of target point pn as residual_radius of target point pn. Intra predictoruses a horizontal angle of the reference point as a predicted value of the horizontal angle of target point pn. Intra predictormay calculate a difference between the predicted value and the horizontal angle of target point pn and calculate a remainder of division of the calculated difference by Δφ as residual_phi of target point pn. That is, intra predictormay calculate a difference between offset φn in horizontal angle from reference position rm to target point pn and an offset in horizontal angle from a reference position (e.g., A1) of the reference point to the reference point (e.g., a1) as residual_phi of target point pn.

100 200 109 109 In the above-described example, the number of points retained for each scan line is three. The number may be any number that is common to three-dimensional data encoding deviceand three-dimensional data decoding device. The intra prediction can be carried out when the number is one. In the case where the number is one, notification of pred_mode may be omitted. Intra predictormay retain an offset in horizontal angle from the reference position to a processed point as information on a horizontal angle of the point and use the offset as the predicted value. In this case, intra predictormay calculate a difference between the offset of target point pn in horizontal angle and the offset of the reference point in horizontal angle as residual_phi.

109 As described above, intra predictorcan reduce values of residual_radius and residual_phi by predicting the information relating to the distance of target point pn and the information relating to the horizontal angle of target point pn, using the information items about the reference positions or the points within the processed range. Therefore, it may be possible to improve coding efficiency.

7 FIG. 7 FIG. Next, another example of a method for selecting a context in the arithmetic encoding process of next_row_flag will be described. Hereinafter, an example of a method for selecting a context to be used in arithmetic encoding of next_row_flag while referring to information on a reference frame that is an encoded frame different from a target frame including a target point.is a diagram for describing the selection of the context.is a diagram illustrating an example of the reference frame.

104 For example, when the target frame is an inter-encoded frame for which inter-encoding is used, entropy encoderselects the context for next_row_flag by the following method.

It should be noted that a frame is point cloud data that expresses a certain three-dimensional space (an object) at a certain time point. The reference frame typically is point cloud data that indicates the same three-dimensional space as the target frame at a time point different from a time point of the target frame. The reference frame need not necessarily be different in time point from the target frame. The reference frame may be point cloud data that expresses a three-dimensional space different from the target frame at the same time point as the time point of the target frame.

7 FIG. 6 FIG. 7 FIG. 104 100 200 illustrates an example of a reference frame of the case of arithmetic encoding of next_row_flag corresponding to reference position rm that target point pn included in the target frame illustrated inrefers to. Entropy encoderselects a context to be used in arithmetic encoding of next_row_flag, using an information item about reference position Col indicated in. For example, reference position Col is a reference position that is the closest to reference position rm of reference positions included in the reference frame. It should be noted that a method for selecting reference position Col is not limited to this. Any technique that enables both three-dimensional data encoding deviceand three-dimensional data decoding deviceto select reference position Col uniquely may be used.

For example, reference position Col may be a reference position that is on a scan line having the same row_pos as reference position rm and has the same column_pos as reference position rm, out of the reference positions included in the reference frame. In other words, reference position Col may be a reference position that has the same values of row_pos and column_pos as values of row_pos and column_pos of reference position rm.

7 FIG. illustrates an example in which there is one point referring to reference position Col. However, there may be no point referring to reference position Col, or there may be points referring to reference position Col. The case where there is no point referring to reference position Col is the case where there is one next_row_flag corresponding to reference position Col, and where a value of next_row_flag is 1. The case there are points referring to reference position Col is the case where there are a plurality of next_row_flags corresponding to reference position Col and having a value of 0. In this case also, reference position Col is determined in the above-described manner.

104 104 6 FIG. When the target frame is the inter-encoded frame, entropy encodermay further apply selection of a context using reference position Col in addition to the selection of a context using the information items about reference positions in the target frame, which is described with reference to. In other words, entropy encodermay select a context, using both the information items about encoded reference positions in the target frame and an information item about reference position Col.

104 Specifically, as a method for selecting a context using reference position Col, entropy encodermay select a context to be used in arithmetic encoding of next_row_flag corresponding to reference position rm in accordance with whether at least one or more next_row_flag corresponding to reference position Col is 0.

104 104 104 104 104 7 FIG. 7 FIG. Alternatively, entropy encodermay select a context using information items about reference positions included in region A that includes reference position Col and the vicinity of reference position Col. For example, for each of reference positions included in region A (9 reference positions in), entropy encodermay determine whether at least one of next_row_flag corresponding to the reference position is 0 and may select a context based on a result of the determination. Specifically, entropy encodermay select a context in accordance with whether the number of reference positions included in the reference positions included in region A and satisfying the condition described above is larger or smaller than the number of reference positions included in the reference positions included in region A and not satisfying the condition. For example, when the number of the reference positions satisfying the condition is larger than the number of the reference positions not satisfying the condition, entropy encoderselects context 0, and when the number of the reference positions satisfying the condition is smaller than the number of the reference positions not satisfying the condition, entropy encoderselects context 1. In the example illustrated in, the number of the reference positions satisfying the condition is 3, and the number of the reference positions not satisfying the condition is 6. Therefore, the number of the reference positions satisfying the condition is determined to be smaller than the number of the reference positions not satisfying the condition.

104 104 Alternatively, entropy encodermay select a context in accordance with the number (a total sum) of the reference positions satisfying the condition. For example, entropy encodermay select a context in accordance with a value of the number of the reference positions satisfying the condition from among 10 contexts including context 0 to context 9 corresponding to 0 to 9 that are numbers of the reference positions satisfying the condition. It should be noted that the value of the number of the reference positions and the contexts need not be associated in a one-to-one correspondence. One context may be associated with values of numbers of the reference positions.

7 FIG. 7 FIG. 7 FIG. 7 FIG. A shape of region A illustrated inis an example. A range of region A is not limited to this. For example, region A may be a range having the same row_pos as reference position Col. For example, region A may include reference position Col illustrated inand reference positions on the right and left sides of reference position Col, three reference positions in total. Alternatively, region A may be a range having the same column_pos as reference position Col. For example, region A may include reference position Col illustrated inand reference positions on the upper and lower sides of reference position Col, three reference positions in total. Alternatively, region A may include reference position Col illustrated inand reference positions on the upper, lower, right and left sides of reference position Col, five reference positions in total.

104 104 104 6 FIG. Entropy encodermay switch between the method for selecting a context in which an information item about reference position Col is referred to and the method for selecting a context in which information items about encoded reference positions in the target frame are referred to described with reference toin accordance with whether the target frame is an inter-encoded frame or an intra-encoded frame for which intra-encoding is used. That is, when the target frame is an inter-encoded frame, entropy encodermay use the method for selecting a context in which an information item about reference position Col is referred to, and when the target frame is an intra-encoded frame, entropy encodermay use the method for selecting a context in which information items about encoded reference positions in the target frame are referred to. The switching need not be performed only on a per frame basis.

100 200 The switching may be performed, for example, on a per reference position basis or on a per set of reference positions basis. In this case, for example, three-dimensional data encoding devicegenerates, on such a basis, a flag that indicates which of the intra prediction and the inter prediction is to be applied, and stores the flag in a bitstream. Three-dimensional data decoding devicemay refer to the flag and switch, on such a basis, between the method for selecting a context in which an information item about reference position Col is referred to and the method for selecting a context in which information items about encoded reference positions in the target frame are referred to.

100 As described above, by selecting a context to be used in arithmetic encoding of next_row_flag, using information items about reference positions in a processed frame, three-dimensional data encoding devicecan switch among contexts more finely. Accordingly, it may be possible to further improve coding efficiency.

Next, a method for inter prediction of a distance and a horizontal angle will be described. The inter prediction is a process of encoding a target point included in a target frame with reference to a reference frame that is an encoded frame different from the target frame. Specifically, one or more encoded reference points included in the reference frame are used to generate a predicted value, and a difference (residual) between information on the target point and the predicted value is calculated. The reference frame typically is point cloud data that indicates the same three-dimensional space as the target frame at a time point different from a time point of the target frame. The reference frame need not necessarily be different in time point from the target frame. The reference frame may be point cloud data that expresses a three-dimensional space different from the target frame at the same time point as the time point of the target frame.

8 FIG. 8 FIG. 8 FIG. 6 FIG. is a diagram for describing an example of the method for inter prediction that predicts information relating to a distance of target point pn and information relating to a horizontal angle of target point pn.is a diagram illustrating an example of the reference frame.also illustrates an example of a reference frame for target point pn included in the target frame illustrated in.

112 112 8 FIG. When a frame to be encoded is an inter-encoded frame, inter predictorcalculates a predicted value of the information relating to the distance of target point pn and a predicted value of the information relating to the horizontal angle of target point pn, using information on the reference frame. For example, inter predictorcalculates the predicted values, using information on reference point RefPt included in the reference frame illustrated in.

100 200 For example, reference point RefPt is a point that is the closest in position to reference position rm of points included in the reference frame. Alternatively, reference point RefPt is a point associated with a reference position that is the closest in position to reference position rm of reference positions included in the reference frame. It should be noted that a method for determining reference point RefPt may be a method that enables both three-dimensional data encoding deviceand three-dimensional data decoding deviceto select reference point RefPt uniquely.

112 112 112 112 112 8 FIG. For example, inter predictorselects a reference position that is on a scan line having the same row_pos as reference position rm and has the same column_pos as reference position rm, from among the reference positions included in the reference frame. In the case where there is one point that refers to such a reference position as in the example illustrated in, inter predictordetermines the point as reference point RefPt. In the case where there are points that refer to such a reference position, inter predictormay determine a point closest to the reference position as reference point RefPt. In the case there is no point that refers to such a reference position, inter predictormay determine, as reference point RefPt, a point that is the closest to the reference position of points on a scan line having the same row_pos as the reference position. Alternatively, inter predictorneed not perform the inter prediction for the information on target point pn, determining that there is no point to be referred to.

112 112 Referring to the information on reference point RefPt, inter predictorcalculates the predicted values of the distance and the horizontal angle. For example, inter predictoruses a distance of reference point RefPt as a predicted value and calculates a difference (residual component) between the distance of target point pn and the predicted value as residual_radius of target point pn.

112 112 Inter predictoralso uses a horizontal angle of reference point RefPt as a predicted value and calculates a difference between the horizontal angle of target point pn and the predicted value. Furthermore, inter predictormay calculate a remainder of division of the calculated difference by Δφ as residual_phi of target point pn.

112 o Inter predictormay use an offset in horizontal angle from reference position Col to reference point RefPt as a predicted value and calculate a difference between the predicted value and offset φn in horizontal angle from reference position rm to target point pn as residual_phi.

100 100 200 6 FIG. 6 FIG. Three-dimensional data encoding devicemay selectively use the method for inter prediction and the method for intra prediction described with reference to. For example, three-dimensional data encoding devicestores mode information that indicates whether to use the inter prediction in syntax information corresponding to point pn included in a bitstream. According to the mode information, three-dimensional data decoding devicecan switch between the method for intra prediction illustrated inand the method for inter prediction.

100 200 When the target frame is an intra-encoded frame, three-dimensional data encoding devicemay omit storing the mode information. In addition, the inter prediction may be assigned to one of prediction modes indicated by pred_mode. In this case, three-dimensional data decoding devicecan switch between the method for intra prediction and the method for inter prediction according to pred_mode.

100 As described above, three-dimensional data encoding devicecan further reduce the value of residual_radius and the value of residual_phi by predicting the information relating to the distance of target point pn and the information relating to the horizontal angle of target point pn, using the information items about points in a processed frame. Therefore, it may be possible to improve coding efficiency.

9 FIG. Next, a flow of the arithmetic encoding process of next_row_flag will be described.is a flowchart illustrating an example of a procedure of the arithmetic encoding process of next_row_flag.

104 101 6 FIG. First, entropy encodersets ctx_intra in accordance with a value or a related parameter of processed next_row_flag in the target frame, using the method described with reference to, for example (S). ctx_intra is information for designating a context to be used in the arithmetic encoding of next_row_flag. The related parameter is, for example, row_pos or column_pos of a reference position corresponding to such next_row_flag.

104 102 100 200 Next, entropy encoderdetermines whether an encoding mode (a decoding mode) used in encoding of a reference position or a three-dimensional point corresponding to next_row_flag to be processed is an intra mode in which the intra prediction is used or an inter mode in which the inter prediction is used (S). Which of the intra mode and the inter mode is used is determined by, for example, three-dimensional data encoding device, and information indicating the determined encoding mode is stored in a bitstream. Referring to the information, three-dimensional data decoding devicedetermines the encoding mode used in encoding of the reference position or the three-dimensional point corresponding to next_row_flag to be processed.

102 104 103 When the intra mode is used (Intra in S), entropy encoderselects a context in accordance with ctx_intra and performs the arithmetic encoding of next_row_flag, using the selected context (S).

102 104 104 7 FIG. On the other hand, when the inter mode is used (Inter in S), entropy encoderfirst selects at least one (one or a plurality of) reference position in a processed frame by the method described with reference toor the like in accordance with a horizontal angle and an elevation angle (or row_pos and column_pos), or the like of a reference position corresponding to next_row_flag to be processed (S).

104 105 Next, entropy encodersets ctx_inter that is information for designating a context to be used in the arithmetic encoding of next_row_flag in accordance with the value of next_row_flag corresponding to the selected at least one of reference positions in the processed frame or the related parameter of such next_row_flag (e.g., row_pos or/and column_pos of the reference position corresponding to such next_row_flag) (S).

104 106 Next, entropy encoderselects a context in accordance with ctx_intra and ctx_inter and performs the arithmetic encoding of next_row_flag, using the selected context (S).

It should be noted that ctx_intra and ctx_inter may be information items of which the number of combinations does not exceed the number of contexts and each of which can designate a context uniquely. For example, ctx_intra and ctx_inter are each expressed with an integer being greater than or equal to 0.

9 FIG. From, a context is selected using both ctx_intra and ctx_inter when the inter mode is used. However, a context may be selected using only ctx_inter.

104 As described above, entropy encodercan select a context in the arithmetic-encoding in accordance with whether an encoding mode for the reference position corresponding to next_row_flag to be processed is the intra mode or the inter mode. Accordingly, it may be possible to further improve coding efficiency of an inter-encoded frame.

100 100 Three-dimensional data encoding devicemay store information that gives instructions about whether to forbid (permit) switching among contexts with reference to information on a processed frame in a frame header of a target frame or a header higher than the frame header. When the information gives the instructions about the forbiddance, three-dimensional data encoding deviceneed not use ctx_inter even when an encoding mode of a reference position corresponding to next_row_flag to be processed is the inter mode and may select a context in accordance with ctx_intra and perform the arithmetic encoding of next_row_flag, using the selected context. Accordingly, it is possible to eliminate dependency on frames in arithmetic encoding and to select an encoding independently for each frame. Therefore, it may be possible to improve error tolerance and improve concurrency in processing.

100 100 Three-dimensional data encoding devicemay store information for identifying a processed frame on which information is referred to when contexts are switched to one another in a frame header of a target frame or a header higher than the frame header. For example, the information is an index value applied to an individual processed frame. Alternatively, three-dimensional data encoding devicemay refer to a processed frame identified for prediction in the inter mode or may identify a processed frame to be referred to by a predetermined method (e.g., selecting a processed frame closest in time or distance).

201 200 104 100 It should be noted that an arithmetic decoding process by entropy decoderincluded in three-dimensional data decoding deviceis similar to the arithmetic encoding process by entropy encoderincluded in three-dimensional data encoding deviceand corresponds to the arithmetic encoding described above in which the arithmetic encoding is replaced with arithmetic decoding.

200 200 200 111 200 112 10 FIG. A flow of a decoding process of position information in three-dimensional data decoding devicewill be described.is a flowchart of the decoding process of position information by three-dimensional data decoding device. Three-dimensional data decoding devicefirst obtains, from a bitstream, mode information indicating whether to use the inter prediction (S). Next, three-dimensional data decoding devicedetermines whether a decoding mode of a target point is the intra mode or the inter mode in accordance with the decoded mode information (S).

112 200 113 200 200 114 200 115 200 4 FIG. 6 FIG. When the intra mode is used (Intra in S), three-dimensional data decoding deviceobtains prediction information (e.g., pred_mode shown in) from the bitstream (S). Next, in accordance with the obtained prediction information, three-dimensional data decoding deviceperforms the method for intra prediction described with reference to. Specifically, three-dimensional data decoding deviceselects one or more processed points in a target frame as one or more reference points for the intra prediction (S). Three-dimensional data decoding devicedetermines a predicted value, using the selected one or more reference points for the intra prediction (S). Specifically, three-dimensional data decoding devicedetermines the predicted value, using information relating to one or more horizontal angles and distances of the selected one or more reference points, and the like.

112 200 200 116 200 117 200 8 FIG. On the other hand, when the inter mode is used (Inter in S), three-dimensional data decoding devicedetermines a predicted value by performing the method for inter prediction described with reference to. Specifically, three-dimensional data decoding deviceselects one or more points in a decoded frame as one or more reference points for the inter prediction in accordance with a reference position corresponding to the target point (S). Next, three-dimensional data decoding devicedetermines the predicted value, using the selected one or more reference points for the inter prediction (S). Specifically, three-dimensional data decoding devicedetermines the predicted value, using information relating to one or more horizontal angles and distances of the selected one or more reference points, and the like.

115 117 200 118 200 115 117 119 After step Sor S, three-dimensional data decoding deviceobtains residual information (e.g., residual_radius and residual_phi) from the bitstream (S). Next, three-dimensional data decoding devicereproduces information relating to a distance and a horizontal angle of the target point from the residual information and the predicted value that is determined in step Sor S(S).

200 113 200 When the target frame is an intra-encoded frame, three-dimensional data decoding devicemay omit the determination as to whether the decoding mode is the intra mode or the inter mode and may always perform the intra prediction (the process of step Sand subsequent steps). In addition, the inter prediction may be assigned to one of values indicated by pred_mode. In this case, three-dimensional data decoding devicemay decode pred_mode before the determination as to whether the decoding mode is the intra mode or the inter mode and determine whether the decoding mode is the intra mode or the inter mode based on decoded pred_mode.

100 100 100 100 10 FIG. In the encoding process, three-dimensional data encoding deviceperforms a prediction process according to a procedure corresponding to the processing procedure illustrated in. For example, three-dimensional data encoding deviceperforms both the intra prediction and the inter prediction, derives residual information items based on predicted values calculated in the intra prediction and the inter prediction, selects, from between the intra prediction and the inter prediction, a prediction method that gives a smaller code amount necessary for encoding the corresponding residual information item, and generates mode information, prediction information, and the residual information item in accordance with a result of the selection. In addition, three-dimensional data encoding devicegenerates a bitstream including the mode information, the prediction information, and the residual information item. The three-dimensional data encoding devicemay store the prediction information in the bitstream only when the mode information is the intra mode.

100 As described above, three-dimensional data encoding devicecan further reduce the value of residual_radius and the value of residual_phi by predicting the information relating to the distance of the target point and the information relating to the horizontal angle of the target point while switching between the intra prediction and the method for inter prediction. Therefore, it may be possible to improve coding efficiency.

100 100 200 100 200 Three-dimensional data encoding devicemay store, in a frame header of the target frame or the like, information for identifying a processed frame on which information is referred to when the inter prediction is performed (e.g., an index value applied to an individual processed frame). Alternatively, three-dimensional data encoding deviceand three-dimensional data decoding devicemay identify a processed frame to be referred to by a predetermined method. For example, three-dimensional data encoding deviceand three-dimensional data decoding devicemay select, as the reference frame, a processed frame that is closest to the target frame in time or distance.

11 FIG. 4 FIG. 5 FIG. 121 122 123 124 As stated above, the three-dimensional data encoding device according to the present embodiment performs the process shown in. The three-dimensional data encoding device determines, from among reference positions, a reference position corresponding to each of three-dimensional points (S). The three-dimensional data encoding device generates first information items (e.g., next_row_flag or next_column_flag) each of which corresponds to a different one of the reference positions and indicates whether a three-dimensional point corresponding to a reference position corresponding to the first information item is present (S). For example, the first information corresponds to at least one of the syntaxes shown inor. The three-dimensional data encoding device selects, from among contexts, a first context for a first information item to be processed that is included in the first information items (S), and arithmetic-encodes the first information item to be processed, using the first context (S).

Accordingly, the three-dimensional data encoding device makes it possible to arithmetic-encode a first information item on a target point, using the first context selected from among the contexts. Accordingly, for example, a context appropriate to each first information item can be used. Therefore, coding efficiency can be improved.

123 For example, in the selecting of the first context (S), the three-dimensional data encoding device selects the first context based on an information item (e.g., at least one of next_row_flag, next_column_flag, row_pos, or column_pos) about a reference position corresponding to an encoded three-dimensional point among the reference positions. Accordingly, the three-dimensional data encoding device makes it possible to appropriately select a context based on the information item about the reference position corresponding to the encoded three-dimensional point.

For example, the information item about the reference position corresponding to the encoded three-dimensional point is a first information item corresponding to the reference position corresponding to the encoded three-dimensional point. Accordingly, the three-dimensional data encoding device makes it possible to appropriately select a context based on the first information item corresponding to the reference position corresponding to the encoded three-dimensional point. Here, a value of the first information item has a high correlation with values of other first information items in a predetermined range in at least one of a temporal direction and a spatial direction. That is, there is a high possibility that, in the vicinity of the first information item, a first information item having the same value appears in the at least one of the temporal direction or the spatial direction. Therefore, by selecting a context, using a first information item corresponding to a reference position corresponding to an encoded three-dimensional point, a context can be used with consideration given to correlation. Thus, it may be possible to reduce an encoding amount.

For example, an information item about positions of the three-dimensional points expresses each of the positions using a distance component, a first direction component, and a second direction component (e.g., a distance, an elevation angle, and a horizontal angle). Accordingly, the three-dimensional data encoding device makes it possible to improve coding efficiency in encoding an information item about a position that is expressed with the distance component, the first direction component, and the second direction component. For example, each of the positions is expressed in a polar coordinate system.

123 For example, each of the reference positions contains the first direction component (e.g., an elevation angle) and the second direction component (e.g., a horizontal angle). In the selecting of the first context (S), the three-dimensional data encoding device selects the first context based on an information item about a first reference position that corresponds to an encoded three-dimensional point and contains a first direction component having a same value as a first direction component contained in a second reference position corresponding to the first information item to be processed.

Accordingly, the three-dimensional data encoding device makes it possible to select the first context appropriately by referring to an information item about a reference position that includes a first direction component having the same value as a first direction component included in a reference position to be processed. For example, like position information items etc. of three-dimensional points obtained by LiDAR, there is a possibility of improving coding efficiency when a correlation between position information items having the same value of the first direction component (e.g., an elevation angle) is high.

3 FIG. For example, the three-dimensional data encoding device arithmetic-encodes information items corresponding to reference positions in an order based on the first direction component in a processing unit composed of the first information items, the reference positions containing second direction components having a same value. For example, a coding order as shown inis used.

Accordingly, when a reference position preceding the reference position to be processed in a coding order differs from the reference position to be processed in the value of the first direction component, the three-dimensional data encoding device makes it possible to select the first context appropriately by referring to an information item about a reference position that includes a first direction component having the same value as a first direction component included in the reference position to be processed.

For example, the first direction component is a vertical component, the second direction component is a horizontal component, and the first information items are arithmetic-encoded in the order based on the vertical component. Accordingly, there is continuity in presence or absence of points to be processed in the vertical direction. It may be possible to reduce an encoding amount in arithmetic encoding.

For example, each of the reference positions contains the first direction component and the second direction component. The three-dimensional data encoding device further calculates a first difference between a value of the first direction component of a three-dimensional point to be processed and a value of the first direction component contained in a first reference position corresponding to the three-dimensional point to be processed. Additionally, the three-dimensional data encoding device may calculate a difference between a value of the second direction component of the three-dimensional point to be processed and a value of the second direction component contained in the first reference position.

Accordingly, the three-dimensional data encoding device makes it possible to improve coding efficiency by calculating the first difference between the value of the first direction component of the three-dimensional point to be processed and the value of the first direction component of the first reference position.

For example, the three-dimensional data encoding device further calculates a second difference between a value of the first direction component of an encoded three-dimensional point and a value of the first direction component contained in a second reference position corresponding to the encoded three-dimensional point, and calculates a third difference that is a difference between the first difference and the second difference. Accordingly, the three-dimensional data encoding device makes it possible to further improve coding efficiency.

For example, the reference positions correspond to sampling positions in a sensor for generating the three-dimensional points. It should be noted that the reference positions are not limited to this example, and may correspond to, for example, predetermined positions or positions spaced apart at a predetermined distance.

For example, when the three-dimensional data encoding device applies intra prediction to a three-dimensional point to be processed, the three-dimensional data encoding device selects the first context by a first method; and when the three-dimensional data encoding device applies inter prediction to the three-dimensional point to be processed, the three-dimensional data encoding device selects the first context by a second method that is different from the first method. Accordingly, the three-dimensional data encoding device makes it possible to select a context appropriate to each of the intra prediction and the inter prediction. Thus, the three-dimensional data encoding device makes it possible to improve coding efficiency.

For example, in the second method, the first context is selected by reference to a processed frame that is different from a frame to be processed that includes the three-dimensional point to be processed. Accordingly, the three-dimensional data encoding device makes it possible to select a context appropriate to the inter prediction. Thus, the three-dimensional data encoding device makes it possible to improve coding efficiency. For example, the three-dimensional data encoding device includes a processor and memory. Using the memory, the processor performs the above process.

12 FIG. 131 132 The three-dimensional data decoding device according to the present embodiment performs the process shown in. The three-dimensional data decoding device selects, from among contexts, a first context for an encoded first information item to be processed (S), and arithmetic-decodes the encoded first information item to be processed, using the first context, to generate a first information item to be processed (S). Here, encoded first information items including the encoded first information item to be processed are included in a bitstream generated by encoding an information item about positions of three-dimensional points. The encoded first information items are generated by arithmetic-encoding first information items including the first information item to be processed. The first information items each correspond to a different one of reference positions and indicate whether a three-dimensional point corresponding to a reference position corresponding to the first information item is present. For example, the three-dimensional data decoding device may further determine whether the three-dimensional point corresponding to the reference position corresponding to the first information item to be processed, using the first information item to be processed.

Accordingly, the three-dimensional data decoding device makes it possible to arithmetic-decode a first information item on a target point, using the first context selected from among the contexts. Accordingly, for example, a context appropriate to each first information item can be used. Therefore, the three-dimensional data decoding device makes it possible to appropriately arithmetic-decode a bitstream of which coding efficiency is improved.

131 For example, in the selecting of the first context (S), the three-dimensional data decoding device selects the first context based on an information item about a reference position (e.g., at least one of next_row_flag, next_column_flag, row_pos, or column_pos) corresponding to a decoded three-dimensional point among the reference positions. Accordingly, the three-dimensional data decoding device makes it possible to appropriately select a context based on the information item about the reference position corresponding to the decoded three-dimensional point.

For example, the information item about the reference position corresponding to the decoded three-dimensional point is a first information item corresponding to the reference position corresponding to the decoded three-dimensional point. Accordingly, the three-dimensional data decoding device makes it possible to appropriately select a context based on the first information item corresponding to the reference position corresponding to the decoded three-dimensional point. Here, a value of the first information item has a high correlation with values of other first information items in a predetermined range in at least one of a temporal direction and a spatial direction. That is, there is a high possibility that, in the vicinity of the first information item, a first information item having the same value appears in the at least one of the temporal direction or the spatial direction. Therefore, by selecting a context, using a first information item corresponding to a reference position corresponding to a decoded three-dimensional point, a context can be used with consideration given to correlation. Thus, it may be possible to reduce an encoding amount.

For example, the information item about the positions of the three-dimensional points expresses each of the positions using a distance component, a first direction component, and a second direction component (e.g., a distance, an elevation angle, and a horizontal angle). Accordingly, the three-dimensional data decoding device makes it possible to decode an information item about a position that is expressed with the distance component, the first direction component, and the second direction component, from a bitstream of which coding efficiency is improved. For example, each of the positions is expressed in a polar coordinate system.

131 For example, each of the reference positions contains the first direction component (e.g., an elevation angle) and the second direction component (e.g., a horizontal angle). In the selecting of the first context (S), the three-dimensional data decoding device selects the first context based on an information item about a first reference position that corresponds to a decoded three-dimensional point and contains a first direction component having a same value as a first direction component contained in a second reference position corresponding to the first information item to be processed.

Accordingly, the three-dimensional data decoding device makes it possible to select the first context appropriately by referring to an information item about a reference position that includes a first direction component having the same value as a first direction component included in a reference position to be processed.

For example, the three-dimensional data decoding device arithmetic-decodes first information items corresponding to reference positions in an order based on the first direction component in a processing unit composed of the first information items, the reference positions containing second direction components having a same value. For example, the three-dimensional data decoding device calculates an information item about positions of three-dimensional points, using reference positions and first information items generated by arithmetic-decoding encoded first information items.

Accordingly, when a reference position preceding the reference position to be processed in a coding order differs from the reference position to be processed in the value of the first direction component, the three-dimensional data decoding device makes it possible to select the first context appropriately by referring to an information item about a reference position that includes a first direction component having the same value as a first direction component included in the reference position to be processed.

For example, the first direction component is a vertical component, the second direction component is a horizontal component, and the first information items are arithmetic-decoded in the order based on the vertical component. Accordingly, there is continuity in presence or absence of points to be processed in the vertical direction. It may be possible to reduce an encoding amount.

For example, each of the reference positions contains the first direction component and the second direction component, and the first information item to be processed indicates a first difference between a value of the first direction component of a three-dimensional point to be processed and a value of the first direction component contained in a first reference position corresponding to the three-dimensional point to be processed. Accordingly, the three-dimensional data decoding device makes it possible to appropriately decode a bitstream of which coding efficiency is improved.

For example, each of the reference positions contains the first direction component and the second direction component, the first information item to be processed indicates a third difference that is a difference between a first difference and a second difference, the first difference is a difference between a value of the first direction component of the three-dimensional point to be processed and a value of the first direction component contained in a first reference position corresponding to the three-dimensional point to be processed, and the second difference is a difference between a value of the first direction component of a decoded three-dimensional point and a value of the first direction component contained in a second reference position corresponding to the decoded three-dimensional point. Accordingly, the three-dimensional data decoding device makes it possible to appropriately decode a bitstream of which coding efficiency is improved.

For example, the reference positions correspond to sampling positions in a sensor for generating the three-dimensional points. It should be noted that the reference positions are not limited to this example, and may correspond to, for example, predetermined positions or positions spaced apart at a predetermined distance.

For example, when the three-dimensional data decoding device applies intra prediction to a three-dimensional point to be processed, the three-dimensional data decoding device selects the first context by a first method; and when the three-dimensional data decoding device applies inter prediction to the three-dimensional point to be processed, the three-dimensional data decoding device selects the first context by a second method that is different from the first method. Accordingly, the three-dimensional data decoding device makes it possible to select a context appropriate to each of the intra prediction and the inter prediction. Thus, the three-dimensional data decoding device makes it possible to improve coding efficiency.

For example, in the second method, the first context is selected by reference to a processed frame that is different from a frame to be processed that includes the three-dimensional point to be processed. Accordingly, the three-dimensional data decoding device makes it possible to select a context appropriate to the inter prediction. Thus, the three-dimensional data decoding device makes it possible to improve coding efficiency.

For example, the three-dimensional data decoding device includes a processor and memory. Using the memory, the processor performs the above process.

A three-dimensional data encoding device, a three-dimensional data decoding device, and the like according to the embodiments of the present disclosure have been described above, but the present disclosure is not limited to these embodiments.

Note that each of the processors included in the three-dimensional data encoding device, the three-dimensional data decoding device, and the like according to the above embodiments is typically implemented as a large-scale integrated (LSI) circuit, which is an integrated circuit (IC). These may take the form of individual chips, or may be partially or entirely packaged into a single chip.

Such IC is not limited to an LSI, and thus may be implemented as a dedicated circuit or a general-purpose processor. Alternatively, a field programmable gate array (FPGA) that allows for programming after the manufacture of an LSI, or a reconfigurable processor that allows for reconfiguration of the connection and the setting of circuit cells inside an LSI may be employed.

Moreover, in the above embodiments, the structural components may be implemented as dedicated hardware or may be realized by executing a software program suited to such structural components. Alternatively, the structural components may be implemented by a program executor such as a CPU or a processor reading out and executing the software program recorded in a recording medium such as a hard disk or a semiconductor memory.

The present disclosure may also be implemented as a three-dimensional data encoding method, a three-dimensional data decoding method, or the like executed by the three-dimensional data encoding device, the three-dimensional data decoding device, and the like.

Also, the divisions of the functional blocks shown in the block diagrams are mere examples, and thus a plurality of functional blocks may be implemented as a single functional block, or a single functional block may be divided into a plurality of functional blocks, or one or more functions may be moved to another functional block. Also, the functions of a plurality of functional blocks having similar functions may be processed by single hardware or software in a parallelized or time-divided manner.

Also, the processing order of executing the steps shown in the flowcharts is a mere illustration for specifically describing the present disclosure, and thus may be an order other than the shown order. Also, one or more of the steps may be executed simultaneously (in parallel) with another step.

A three-dimensional data encoding device, a three-dimensional data decoding device, and the like according to one or more aspects have been described above based on the embodiments, but the present disclosure is not limited to these embodiments. The one or more aspects may thus include forms achieved by making various modifications to the above embodiments that can be conceived by those skilled in the art, as well forms achieved by combining structural components in different embodiments, without materially departing from the spirit of the present disclosure.

The present disclosure is applicable to a three-dimensional data encoding device and a three-dimensional data decoding device.